Gemini Provider

Integrate Google's Gemini API with Drupal through the AI module. This provider brings the full power of Google's Gemini models to your Drupal site - from chat and embeddings to image generation, speech synthesis, and more.

Supported Operations

• Chat — Text generation and conversation with streaming support. Compatible with the Gemini chat models available to your API key.
• Embeddings — Generate vector embeddings from text and images for semantic search and RAG pipelines.
• Text-to-Image — Generate images from text prompts using Gemini's native image generation with configurable aspect ratios.
• Text-to-Speech — Convert text to natural-sounding audio with multiple voice presets (Kore, Puck, Charon, Fenrir, Aoede, and more).
• Speech-to-Text — Transcribe audio files to text with language detection support.
• Structured Output — Get responses in structured JSON format using schema definitions.
• Tools / Function Calling — Let Gemini models call your custom functions for agentic workflows.

Key Features

• Capability declaration for seamless integration with the AI module's provider system
• Token usage tracking for chat
• Streamed chat responses for real-time output
• Configurable generation parameters (temperature, top-p, top-k, max tokens, stop sequences)